A&A 468, L25–L28 (2007) Astronomy DOI: 10.1051/0004-6361:20077294 & c ESO 2007 Astrophysics
Letter to the Editor
The water ice rich surface of (145453) 2005 RR43: a case for a carbon-depleted population of TNOs?
N. Pinilla-Alonso1, J. Licandro2,3, R. Gil-Hutton4, and R. Brunetto5,6
1 Fundación Galileo Galilei & Telescopio Nazionale Galileo, PO Box 565, 38700, S/C de La Palma, Tenerife, Spain e-mail: [email protected] 2 Isaac Newton Group, 38700 Santa Cruz de La Palma, Tenerife, Spain 3 Instituto de Astrofísica de Canarias, c/Vía Láctea s/n, 38205 La Laguna, Tenerife, Spain 4 Complejo Astronómico El Leoncito (Casleo) and San Juan National University, Av. España 1512 sur, J5402DSP, San Juan, Argentina 5 Dipartimento di Fisica, Università del Salento, via Arnesano, 73100 Lecce, Italy 6 INAF-Osservatorio Astrofisico di Catania, via S. Sofia 78, 95123 Catania, Italy Received 13 February 2007 / Accepted 23 April 2007
ABSTRACT
Context. Recent results suggest that there is a group of trans-Neptunian objects (TNOs) (2003 EL61 being the biggest member), with surfaces composed of almost pure water ice and with very similar orbital elements. These objects provide exciting laboratories for the study of the processes that prevent the formation of an evolved mantle of organics on the surfaces of the bodies in the trans-Neptunian belt (TNb). Aims. We study the surface composition of another TNO that moves in a similar orbit, (145453) 2005 RR43, and compare it with the surface composition of the other members of the group. Methods. We report visible and near-infrared spectra in the 0.53−2.4 µm spectral range, obtained with the 4.2 m William Herschel Telescope and the 3.58 m Telescopio Nazionale Galileo at the “Roque de los Muchachos” Observatory (La Palma, Spain). Scattering models are used to derive information about its surface composition. We also measure the depth D of the water ice absorption bands and compare with those of the other members of the group. Results. The spectrum of 2005 RR43 is neutral in color in the visible and dominated by very deep water ice absorption bands in the near infrared (D = 70.3 ± 2.1% and 82.8 ± 4.9% at 1.5 µm and 2.0 µm respectively). It is very similar to the spectrum of the group of TNOs already mentioned. All of them present much deeper water ice absorption bands (D > 40%) than any other TNO except Charon. Scattering models show that its surface is covered by water ice, a significant fraction in crystalline state with no trace (5% upper limit) of complex organics. Possible scenarios to explain the existence of this population of TNOs are discussed: a giant collision, an originally carbon depleted composition, or a common process of continuous resurfacing. Conclusions. 2005 RR43 is member of a group, may be a population, of TNOs clustered in the space of orbital parameters that show abundant water ice and no signs of complex organics and which origin needs to be further investigated. The lack of complex organics in their surfaces suggests a significant smaller fraction of carbonaceous volatiles like CH4 in this population than in “normal” TNOs. A carbon depleted population of TNOs could be the origin of the population of carbon depleted Jupiter family comets already noticed by A’Hearn et al. (1995). Key words. Kuiper Belt – solar system: formation
1. Introduction (Brown et al. 1999) also have surface composition similar to Charon. During last year the spectra of five other objects were Spectroscopic and spectrophotometric studies show that published, revealing that their surfaces are also covered by fresh about 70% of TNOs present a mantle of complex organics on water ice: (136108) 2003 EL61 (Trujillo et al. 2007), its biggest their surfaces (Brunetto et al. 2006). Long term processing by satellite S/2005 (136108) 1 (Barkume et al. 2006) and during high energy particles and solar radiation on icy bodies, induces the review process of this paper 2003 OP32, 1995 SM55 and the formation of organic species in their outer layers, resulting in 2005 RR43 (Brown et al. 2007). The spectra of these TNOs show a mantle that covers the unprocessed original ices (Moore et al. all the same characteristics, they are neutral and featureless in the 1983; Johnson et al. 1984; Strazzulla & Johnson 1991). Until visible and show strong water ice absorption bands in the near recently, the only case of a TNO with a surface covered basically infrared. All these TNOs, except Charon, are located in a narrow by a thick layer of water ice was Charon (Buie et al. 1987; region of the orbital parameters space (41.6 < a < 43.6AU, Marcialis et al. 1987), and it has been considered an intringuing 25.8 < i < 28.2deg,0.10 < e < 0.19). The existence of a pop- case because of the need of a resurfacing mechanism like ulation of TNOs with Charon-like surfaces and similiar orbital cryovolcanism or collisons with micro-meteorites (Brown 2002; parameters needs to be explained as it can have a strong impact Cruikshank 1998). Recently, it has been showed that (55636) in the knowledge of the trans-neptunian belt formation theories 2002 TX300 (Licandro et al. 2006) and (13308) 1996 TO66 and/or resurfacing mechanisms.
Article published by EDP Sciences and available at http://www.aanda.org or http://dx.doi.org/10.1051/0004-6361:20077294 L26 N. Pinilla-Alonso et al.: The surface of (145453) 2005 RR43: a case for a carbon-depleted population of TNOs?
In this paper we present visible and near-infrared spec- troscopy of a member of this group, (145453) 2005 RR43 (a, e, i = 43.06 AU, 0.14, 28.54 deg) and derive compositional information using scattering models. Finally we describe differ- ent scenarios that can explain the existence of this population of TNOs.
2. Observations and analysis of the spectrum
We obtained the visible spectrum of 2005 RR43 with the 4.2 m William Herschel telescope (WHT) and the near-infrared spec- trum with the 3.58 m “Telescopio Nazionale Galileo” (TNG) both at the “Roque de los Muchachos” Observatory (Canary Islands, Spain). The visible spectrum (0.35−0.95 µm) was obtained on
2006 Sep. 23.18 UT using the low resolution grating (R158R, Fig. 1. Spectrum of 2005 RR43, normalized at 0.55 µm, together with with a dispersion of 1.63 Å/pixel) of the double-armed spectro- the spectra of the other members of the group shifted in the vertical axis graph ISIS at WHT, and a 4 slit width oriented at the parallactic for clarity. References in Table 1. angle. The tracking was at the TNO proper motion. Three 900 s exposure time spectra were obtained by shift- ing the object by 10 in the slit to better correct the fringing. the depth of the bands, D, with respect to the continuum of the Calibration and extraction of the spectra were done using IRAF = − / and following standard procedures (Massey et al. 1992). TNO spectrum as D 1 Rb Rc,whereRb is the reflectance in the spectra were averaged and the reflectance spectrum was obtained center of the band and Rc is the reflectance of the continuum at 1.2 µm. For 2005 RR43, D is 70.3 ± 2.1% and 82.8 ± 4.9% by dividing the spectrum of the TNO by the spectrum of the so- µ µ lar analogue star Hyades 64 obtained the same night just before at 1.5 m and 2.0 m respectively, beeing the deepest water ice and after the observation of the TNO at a similar airmass. Final absorption bands ever observed in a TNO. spectrum presented in Fig. 1 was smoothed using a smoothing Thus, we conclude that the surface of 2005 RR43 is com- box-car of 15 pixels to improve the S/N. posed by a large fraction of large sized (or a thick layer of) wa- The near-infrared spectrum was obtained on ter ice particles, and none or a very small fraction of complex 2006 Sep. 29.15 UT, using the low resolution spectroscopic organics or silicates. mode of NICS (Near-Infrared Camera and Spectrometer) at the This is confirmed by scattering models. We use the TNG based on an Amici prism disperser. This mode yields a simple one-dimensional geometrical-optics formulation by complete 0.8−2.4 µm spectrum. A 1.5 slit width corresponding Shkuratov et al. (1999), to obtain information about the surface to a spectral resolving power R 34 quasi-constant along the composition. spectrum was used. The observing and reduction procedures It is important to determine if water ice is in amorphous or were as described in Licandro et al. (2002). The total “on crystalline state as it can be indicative of resurfacing processes. object” exposure time was 11 160 s. Crystalline water ice can be easily identified by an absorption To correct for telluric absorption and to obtain the relative band at 1.65 µm. This band has been detected in the spec- reflectance, solar analogue star Hyades 64 and two G-2 Landolt trum of TNOs Charon (Brown & Calvin 2000) and 2003 EL61 stars (115−271 and 98−978, Landolt 1992) were observed at dif- (Trujillo et al. 2007), the only TNOs that have spectra similar ferent airmasses during the night. The reflectance spectrum of to 2005 RR43 and sufficiently high signal-to-noise to see this the TNO was obtained using all the SA stars, averaged and then band. Unfortunately, the relatively poor S/N of our spectrum normalized to fit the visible one using the overlapping spectral does not allow us to clearly detect this feature. − µ region between 0.85 0.95 m. The combined visible and near- So we first tried to model 2005 RR43 spectrum with pure infrared (VNIR) spectrum is presented in Fig. 1 together with amorphous water ice, but all the tests we made produced a band the spectra of the other members of the group. at 1.5 µm narrower than that seen in the spectrum (see Fig. 2). The VNIR spectrum reveals three important characteris- We then tried with pure crystalline water ice obtaining a better tics: (a) the visible is featureless within the S/N. There is no fit. Finally we used an intimate mixture of crystalline and amor- clear evidence of any absorption reported for other TNOs (e.g. phous water ices obtaining even better results and improving the Lazzarin et al. 2004; Fornasier et al. 2004a); (b) it is neutral fit above 2 µm, although this part of the spectrum is too noisy (the spectral slope, computed between 0.53 and 1.00 µm, is to be a significant constraint. We thus conclude that water ice S = 0.4 ± 1%/1000 Å), far from the more characteristical red in the surface of 2005 RR43 is, at least in a significant fraction, slopes in the TNb (Fornasier et al. 2004b); (c) it presents two crystalline. deep absorption bands centered at 1.5 and 2.0 µm, indicative of On the other hand, all these models allow a small percent- water ice. age of minor components such as amorphous carbons or silicates The flat visible spectrum is indicative of the lack (or very low (e.g. Olivine), up to 5%. We notice that the S/N of the spectrum abundance) of complex organic and/or silicates in the surface of and the model itself do not allow us to make a detailed study this TNO. of the surface mineralogy. An important parameter to include, The depth of the water ice bands is a good indicator of its that could help to better constraint the models and, in particu- abundance on the surface of icy objects, it is also a reasonably lar, determine the abundance of minor dark constituents, is the good marker of the size of the icy particles and of the contami- albedo of the TNO, as even a small amount of them can darken nation by non-ice components (Clark et al. 1984). We computed the surface significantly. N. Pinilla-Alonso et al.: The surface of (145453) 2005 RR43: a case for a carbon-depleted population of TNOs? L27
Table 1. H: absolute magnitud; S’: spectral gradient; D: absorption at 1.5 µm. References: 1. Brown & Calvin ( 2000), 2. Fink & DiSanti (1988), 3. The online updated database from Hainaut & Delsanti (2002), 4. Brown et al. (2007), 5. Brown et al. (1999), 6. Noll et al. ( 2000), 7. Licandro et al. (2006), 8. Tegler et al. (2007), 9. Trujillo et al. (2007), 10. Barkume et al. (2006), 11. This work.
Object HV S D Ref Charon 0.9 –0.6 41.5 ± 3.4% (1, 2) 1995 SM55 4.8 2.4 70.2 ± 5.9% (3, 4) 1996 TO66 4.5 4.5 59.1 ± 1.9% (3, 5, 6) 2002 TX300 3.3 1 64.1 ± 3.9% (7) 2003 OP32 4.1 –1.1 64.2 ± 2.5% (4, 8) (136108) 2003 EL61 0.2 –0.4 40.7 ± 2.1% (8, 9) S/2005 (136108) 1 3.5 NA 69.1 ± 1.0% (10) 2005 RR43 4.0 –0.4 70.3 ± 2.1% (11) Fig. 2. Best fits of the spectrum with the Shkuratov approximation (in- timate mixtures of amorphous water ice (AWI), crystalline water ice (CWI) and olivine (O)). Model 1: 95% AWI (70 µm)5% O (30 µm); (Gil-Hutton 2002). The material sublimated by the collision can model 2: 95% CWI (50 µm) + 5% O (30 µm); model 3: 55% AWI be globally redeposited over the TNO on a timescale of tens of µ + µ + µ µ (60 m) 40% CWI (70 m) 5% O (30 m). The width of 1.5 m hours (Stern 2002) and the low vertical diffusion velocity of the band is best fitted with crystalline water ice. ice ensures that large particles can be formed with a high effi- ciency while being downward transported and deposited on the 3. Discussion and conclusions surface. Brown et al. (2007) proposed that the TNOs listed in Table 1 (but Charon) are fragments produced by a catastrophic To date, about 40 TNOs have been observed spectroscopically collision suffered by 2003 EL61. This would explain the rapid in the near-infrared. Among them only seven exhibit a spectrum rotation of this large object, the existence of two satellites orbit- similar to that of 2005 RR43. Table 1 summarizes their 1.5 µm ing it, and the clustering of objects in a small region of the TNb band depth (D) and spectral slope S computed as in Sect. 3 with similar surface properties. for 2005 RR43. Notice that all of these TNOs have very deep Although this scenario appears promising, it has several water ice absorptions, D > 40% and S ∼ 0. problems: first, the low mean intrinsic collisional probability Brown et al. (2007) noticed that, excluding Charon, these makes highly improbable a catastrophic collision with a large objects present very similar spectral properties and orbital pa- projectile capable of providing enough kinetic energy to disperse rameters (see Fig. 2 of Brown’s paper). On the other hand, the large fragments to their present positions, even in a TNb more spectra and colors of other TNOs in the neighbourhood of the massive that the present one. Second, it is difficult to model the cluster, e.g. (20000) Varuna (Licandro et al. 2001) or (50000) collisional process to produce such a family working with so Quaoar (Jewitt & Luu 2004), are different from the spectrum of few members and given the large uncertainties in their orbital 2005 RR43 and present the variety of colors and composition ob- elements. Third, the current dispersion in orbital elements of served in other regions of the TNb. Thus, we also conclude that the fragments shows orbital elements expected from a too high objects in Table 1 (except Charon) are part of a population clus- dispersive velocity of 400 m/s. Brown et al. argued that only tered in the space of parameters and with similar surface proper- 2003 EL61 does not fit withing a smaller velocity of 140 m/sand ties different from those of the objects in the neighbourhood. argued that this TNO has large excursions in eccentricity over There are other smaller TNOs, with no photometric or spec- time owing to chaotic diffusion within the 12:7 mean-motion troscopic data published, that according to their orbital param- resonance (MMR) with Neptune. If this is the case, following eters, could be also members of this population like 1995 GJ, the analysis by Nesvorny & Roig (2001) of the 12:7 MMR with 1999 OK4, 1999 RA215, 2003 FB130, 2005 PM21. Photometric Neptune, and cited by Brown et al. (2007), the chaotic diffu- and/or spectroscopic observations are needed. sion enlarge the initial eccentricity of any orbit near its bor- Water ice surfaces with no traces of organics should not be ders in such a way that it becomes Neptune grazing and the common in the TNb assuming that the original chemical compo- body escapes in less than 108 yr, dispersing very fast any ob- sition of all TNOs is very similar: objects composed of abundant ject in its vicinity. But the collision must have happened in the water ice, some molecular ices like CO, CO2,CH4,N2 and sili- distant past, when the belt was far more crowded with large cates. Long term processing by high energy particles and solar objects (Morbidelli 2007). Fourth, if the collision happened in radiation induces the formation of complex organic species in the distant past, the effects of long term processing by high the outer layers of the TNOs resulting in a dark and usually red energy particles and solar radiation should be present and are mantle that covers the unprocessed original ices. Thus, the ex- not. Gil-Hutton ( 2002) shows that the time scale required to istence of a population of TNOs with no signs of organics, in form a black irradiation mantle of carbon residues in a TNO is a very narrow region of the space of orbital parameters is an ∼6 × 108 years. The competition of this effect with resurfacing intriguing fact that needs to be explained. We discuss three dif- due to high frequency impacts will increase the time scale to ferent scenarios that need to be further studied: 109 years while intermediate states would result in red spectra. 1) Collisional family: the destruction of the irradiation man- Moreover the efficiency of this process depends on the presence tle by an energetic collision has been proposed by Licandro of organics on the surface so the absence of an irradiation mantle et al. (2006) to explain the fresh surface of one of the mem- can be explained by the loss of organics or the absence of them in ber of this population, 2002 TX300. Such an impact breaks the original composition. Thus, a possible solution of this prob- the irradiation mantle and produces enough energy to subli- lem is that the loss of organics was produced by the collision and mate a certain amount of ices on the upper layers of the body recondensation process that formed the family. L28 N. Pinilla-Alonso et al.: The surface of (145453) 2005 RR43: a case for a carbon-depleted population of TNOs?
2) Originally carbon depleted population: another possi- References ble scenario is a population of objects originally carbon depleted, A’Hearn, M. F., Millis, R. L., Schleicher, D. G., et al. 1995, Icarus, 118, 223 strongly concentrated in the space of orbital parameters, and in Baffa, C., Comoretto, G., Gennari, S., et al. 2001, A&A, 378, 722 a region that it is dynamically unstable (the hot population). But Barkume, K. M., Brown, M. E., & Schaller, E. L. 2006, ApJ, 640, L87 this has also several problems: why carbon depleted objects were Brown, M. 2002, Ann. Rev. Earth. Planet. Sci., 30, 307 formed in a narrow region of the solar nebula and remained Brown, M., & Calvin, W. 2000, Science, 287, 107 Brown, R., Cruikshank, D., & Pendleton, Y. 1999, ApJ, 519, L101 grouped until present time? However, the existence of a popu- Brown, M. E., Barkume, K. M., Ragozinne, D., et al. 2007, Nature, 446, 294 lation of carbon depleted TNOs is an interesting case, as it could Brunetto, R., Barucci, M., Dotto, M. E., & Strazzulla, G. 2006, ApJ, 644, 650 be the origin of the population of carbon depleted Jupiter fam- Buie, M., Cruikshank, D., Lebofsky, L., et al. 1987, Nature, 329, 522 ily comets already noticed by A’Hearn et al. (1995) and whose Clark, R. N., Brown, R. H., Owensby, P. D., et al. 1984, Icarus, 58, 265 origin remains unknown. Cruikshank, D. 1998, in Solar System Ices, ed. B. Schmitt et al. (Kluwer), 655 Fink, U., & DiSanti, M. A. 1988, AJ, 95, 229 3) Continuous resurfacing process: another possible sce- Fornasier, S., Dotto, E., Barucci, A., et al. 2004a, A&A, 422, 43 nario is that these objects are exposed to a mechanism that re- Fornasier, S., Doressoundiram, A., Tozzi, et al. 2004b, A&A, 421, 353 plenishes their surfaces permanently with fresh material from Gil-Hutton, R. 2002, P&SS, 50, 57 their interiors like cryovolcanism. Anyhow, it is difficult to ex- Hainaut O. R., & Delsanti, A. C. 2002, A&A, 389, 641 ff ffi http://www.sc.eso.org/ ohainaut/MBOSS/ plain why this mechanism a ects so e ciently this group of Jewitt, D. C., & Luu, J. 2004, Nature, 432, 731 TNOs and only this group. Johnson, R., Lanzerotti, L., & Brown, W. 1984, Adv. Space Res. 4, 41 In conclusion, the spectrum of TNO 2005 RR43 in the visible Landolt, A. 1992, AJ, 104, 340 and near-infrared range shows that its surface is covered by large Lazzarin, M., Barucci, M., Boehnhardt, H., et al. 2003, AJ, 125, 1554 water ice grains. Scattering models reveal that the observed wa- Licandro, J., Oliva, E., & Di Martino, M. 2001, A&A, 373, L29 Licandro, J., Ghinassi, F., & Testi, L. 2002, A&A, 388, L9 ter ice is, at least in a significant fraction, crystalline. 2005 RR43 Licandro, J., di Fabrizio, L., Pinilla-ALonso, N., et al. 2006, A&A, 457, 323 spectrum is very similar to those of TNOs Charon, 1996 TO66, Marcialis, R., Rieke, G., & Lebofsky, L. 1987, Science, 237, 1349 2002 TX300, 2003 OP32, 1995 SM55, (136108) 2003 EL61 and Massey, P., Valdes, F., & Barnes, J. 1992, in A User’s Guide to Reducing Slit S/2005 (136108). It also has orbital elements very similar to Spectra with IRAF, http://iraf.noao.edu/iraf/ftp/iraf/docs/spect.ps.Z. those of these last four TNOs, supporting the existence of a pop- Moore, M., Donn, B., Khanna, R., & A’Hearn, M. 1983, Icarus, 54, 388 ulation of TNOs with their surface covered by fresh water ice Morbidelli, A. 2007, Nature, 446, 273 and almost no complex organics. The lack of complex organics Nesvorny, D., & Roig, F. 2001, Icarus, 150, 104 suggests a signficant smaller fraction of carbonaceous volatiles Noll, K. S., Luu, J., & Gilmore, D. 2000, AJ, 119, 970 ff like CH in this population than in “normal” TNOs. Such carbon Shkuratov, Y., Starukhina, L., Ho mann, H., et al. 1999, Icarus, 137, 235 4 Stern, S. A. 2002, AJ, 124, 2297 depleted population of TNOs could be the origin of the popula- Strazzulla, G., & Johnson, R. 1991, in Comets in the Post-Halley era, ed. R. L. tion of carbon depleted Jupiter family comets already noticed by Newburn Jr., M. Neugebauer, & J. Rahe (Netherlands: Kluwer Academic A’Hearn et al. (1995). The origin of this population needs to be Publishers), 243 further investigated. Tegler, S. C., Grundy, W. M., Romanishin, W., et al. 2007, AJ, 133, 526 Trujillo, C. A., Brown, M., Barkume, K. M., et al. 2007, ApJ, 655, 1172 Acknowledgements. We wish to thank Ted Roush for providing optical constants and Humberto Campins for his usefull comments.